Network


Latest external collaboration on country level. Dive into details by clicking on the dots.

Hotspot


Dive into the research topics where Matthew Ager is active.

Publication


Featured researches published by Matthew Ager.


information theory and applications | 2008

Discriminative and generative machine learning approaches towards robust phoneme classification

Jibran Yousafzai; Matthew Ager; Zoran Cvetkovic; Peter Sollich

Robustness of classification of isolated phoneme segments using discriminative and generative classifiers is investigated for the acoustic waveform and PLP speech representations. The two approaches used are support vector machines (SVMs) and mixtures of probabilistic PCA (MPPCA). While recognition in the PLP domain attains superb accuracy on clean data, it is significantly affected by mismatch between training and test noise levels. Classification in the high-dimensional acoustic waveform domain, on the other hand, is more robust in the presence of additive white Gaussian noise. We also show some results on the effects of custom-designed kernel functions for SVM classification in the acoustic waveform domain.


international symposium on information theory | 2011

Combined waveform-cepstral representation for robust speech recognition

Matthew Ager; Zoran Cvetkovic; Peter Sollich

High-dimensional acoustic waveform representations are studied as a front-end for noise robust automatic speech recognition using generative methods, in particular Gaussian mixture models and hidden Markov models. The proposed representations are compared with standard cepstral features on phoneme classification and recognition tasks. While lower error rates are achieved using cepstral features at very low noise levels, the acoustic waveform representations are much more robust to noise. A convex combination of acoustic waveforms and cepstral features is then considered and it achieves higher accuracy than either of the individual representations across all noise levels.


information theory and applications | 2010

High-dimensional linear representations for robust speech recognition

Matthew Ager; Zoran Cvetkovic; Peter Sollich

Phoneme classification is investigated in linear feature domains with the aim of improving the robustness to additive noise. Linear feature domains allow for exact noise adaptation and so should result in more accurate classification than representations involving nonlinear processing and dimensionality reduction. We develop a generative framework for phoneme classification using linear features. We first show results for a representation consisting of concatenated frames from the centre of the phoneme, each containing f frames. As no single f is optimal for all phonemes, we further average over models with a range of values of f. Next we improve results by including information from the entire phoneme. In the presence of additive noise, classification in this framework performs better than an analogous PLP classifier, adapted to noise using cepstral mean and variance normalisation, below 18dB SNR.


IEEE | 2012

Problems of Redundancy in Information and Control Systems (RED), 2012 XIII International Symposium on

Jibran Yousafzai; Matthew Ager; Zoran Cvetkovic; Peter Sollich

Automatic speech recognition (ASR) systems are yet to achieve the level of robustness inherent to speech recognition by the human auditory system. The primary goal of this paper is to argue that exploiting the redundancy in speech signals could be the key to solving the problem of the lack of robustness. This view is supported by our recent results on phoneme classification and recognition in the presence of noise which are surveyed in this paper.


2012 XIII International Symposium on Problems of Redundancy in Information and Control Systems | 2012

Redundancy in speech signals and robustness of automatic speech recognition

Jibran Yousafzai; Zoran Cvetkovic; Matthew Ager; Peter Sollich

Automatic speech recognition (ASR) systems are yet to achieve the level of robustness inherent to speech recognition by the human auditory system. The primary goal of this paper is to argue that exploiting the redundancy in speech signals could be the key to solving the problem of the lack of robustness. This view is supported by our recent results on phoneme classification and recognition in the presence of noise which are surveyed in this paper.


european signal processing conference | 2009

Robust phoneme classification: Exploiting the adaptability of acoustic waveform models

Matthew Ager; Zoran Cvetkovic; Peter Sollich


european signal processing conference | 2008

Towards robust phoneme classification: Augmentation of PLP models with acoustic waveforms

Matthew Ager; Zoran Cvetkovic; Peter Sollich; Bin Yu


Archive | 2013

Phoneme Classification in High-Dimensional Linear Feature Domains.

Matthew Ager; Zoran Cvetkovic; Peter Sollich


arXiv: Computation and Language | 2013

Speech Recognition Front End Without Information Loss

Matthew Ager; Zoran Cvetkovic; Peter Sollich


arXiv: Computation and Language | 2013

A Subband-Based SVM Front-End for Robust ASR

Jibran Yousafzai; Zoran Cvetkovic; Peter Sollich; Matthew Ager

Collaboration


Dive into the Matthew Ager's collaboration.

Top Co-Authors

Avatar
Top Co-Authors

Avatar
Top Co-Authors

Avatar
Top Co-Authors

Avatar

Bin Yu

University of California

View shared research outputs
Researchain Logo
Decentralizing Knowledge